Eric Rozner - ETX.ppt1 A High-Throughput Path Metric for Multi-Hop Wireless Routing Douglas S.J. Couto Daniel Aguayo John Bicket Robert Morris Presented by: Eric Rozner

Eric Rozner - ETX.ppt2 Overview  Hop count metric  Describe ETX  Implementation  Experiments and Comparisons  Related Work  Conclusions  Discussion

Eric Rozner - ETX.ppt3 Hop Count Metric  Paths decided upon by shortest route  Maximizes the distance traveled by each hop Minimizes signal strength -> Maximizes the loss ratio Uses a higher TxPower -> Interference  Possibly many shortest routes  Avoid lossy links?

Eric Rozner - ETX.ppt4 Hop Count vs. “Optimal” x axis: throughput y axis: fraction of pairs with less throughput

Eric Rozner - ETX.ppt5 Hop Count Route Selection

Eric Rozner - ETX.ppt6 Motivation for a Better Routing Metric Bidirectional loss ratesFine-grained choices Intermediate loss rates

Eric Rozner - ETX.ppt7 Potential Ideas (and their cons)  Product of per-link delivery ratios A perfect 2-hop route is viewed as better than a 1-hop route with 10% loss ratio  Throughput of a path’s bottleneck link Same as above  End-to-End delay Changes with network load as interference queue lengths vary… can cause oscillations  Everyone convinced?

Eric Rozner - ETX.ppt8 ETX  The predicted number of data transmissions required to send a packet over a link  The ETX of a path is the sum of the ETX values of the links over that path  Examples: ETX of a 3-hop route with perfect links is 3 ETX of a 1-hop route with 50% loss is 2

Eric Rozner - ETX.ppt9 ETX continued…  Expected probability that a transmission is successfully received and acknowledged is d f x d r d f is forward delivery ratio d r is reverse delivery ratio  Each attempt to transmit a packet is a Bernoulli trial, so…

Eric Rozner - ETX.ppt10 Hooray for ETX!  Based on delivery ratios, which affect throughput  Detects and handles asymmetry by incorporating loss ratios in each direction  Uses precise link loss ratios measurements to make fine-grained decisions between routes Assumes you can measure these ratios precisely  Penalizes routes with more hops, which have lower throughput due to inter-hop interference Assumes loss rates are generally equal over links  Tends to minimize spectrum use, which should maximize overall system capacity (reduce power too) Each node spends less time retransmitting data

Eric Rozner - ETX.ppt11 ETX always the best?

Eric Rozner - ETX.ppt12 Acquiring ETX values  Measured by broadcasting dedicated link probe packets with an average period τ (jittered by ±0.1τ)  Delivery ratio: count(t-w,t) is the # of probes received during window w w/τ is the # of probes that should have been received  Each probe contains this information

Eric Rozner - ETX.ppt13 Implementation and such…  Authors modified DSDV and DSR See paper for details  τ = 1 packet per second, w = 10 sec  Multiple queues (different priorities) Loss-ratio probes, protocol packets, data packets  Are these experiments unfair or unrealistic? In DSDV w/ ETX, route table is a snapshot taken at end of 90 second warm-up period In DSR w/ ETX, source waits additional 15 sec before initiating the route request

Eric Rozner - ETX.ppt14 DSDV Performance One hop Asymmetric ETX inaccurate

Eric Rozner - ETX.ppt15 DSDV and High Transmit Power

Eric Rozner - ETX.ppt16 Packet Size Problems  Less throughput advantage than when data packets are smaller (134 bytes)

Eric Rozner - ETX.ppt17 Packet sizes continued ACK’s smaller than probe packets ETX underestimates ACK delivery ratios -> overestimates total number of transmissions per packet

Eric Rozner - ETX.ppt18 DSR Performance Link-layer transmission feedback disabled Link-layer transmission feedback enabled

Eric Rozner - ETX.ppt19 Related Work  Yarvis et al. [33] propose a path metric which approximates the product of the per-link delivery ratios.  Link hand-shaking to avoid “gray areas” [8], [22]  SNR ratio is also a possible path metric [14] SNR threshold value to filter links discovered by DSR Route Discovery.  See Related Works section of “Comparison of Routing Metrics for Static Multi-Hop Wireless Networks “Comparison of Routing Metrics for Static Multi-Hop Wireless Networks

Eric Rozner - ETX.ppt20 Related Work Continued…  WCETT proposed in MR-LQSR Weighted Cumulative Expected Transmission Time

Eric Rozner - ETX.ppt21 Conclusions  Pros ETX performs better or comparable to Hop Count Metric  Accounts for bi-directional loss rates Can easily be incorporated into routing protocols A lot of other work built off this paper  Cons May not be best metric for all networks  Mobility, Power-limited, Adaptive Rate (multi- rate), Multi-radio Predications of loss ratios not always accurate Experiments (30 sec transfer of small packets) may not complement real-world scenarios

Eric Rozner - ETX.ppt22 Questions?